Multi-Type Feature Fusion Model For Drug-Drug Synergy Prediction

Drug combination therapy is an effective solution in the treatment of cancer and Drug-Drug Synergy (DDS) detection has become a research hotspot in pharmacology. This paper proposes a multi-type feature fusion model for drug synergy prediction, called LGMTF. This model can effectively integrate the topological information and sequence information of the drug to improve the accuracy of the model. Experiment results show that LGMTF outperforms the contrast models for DDS prediction. In addition, the case studies show that some synergistic drug combinations predicted by our method are consistent with the result of clinical experiments.